Michael A. Covington Michael A. Covington, Ph.D.
Senior Research Scientist
Adjunct Professor of Computer Science
Associate Director
Artificial Intelligence Center
The University of Georgia
This is a private web page not hosted or sponsored by the University.
Copyright 2006 Michael A. Covington.
Caching by search engines is explicitly permitted.
To go to the latest entry every day, bookmark
Covington Innovations
Previous months
About this notebook
Google Search site or Web
--- ---

Daily Notebook

--- ---
Astrophotography for the Amateur How to Use a Computerized Telescope Dictionary of Computer and Internet Terms

Popular topics on this page:
The book that launched Ramanujan
Olympus OM-1 battery modification
Olympus OM-10 top cover removal
Olympus OM-10 sticky frame counter
Textbook payola scandal
Credit card industry gets well-deserved flak

For more topics, scroll down, press Ctrl-F to search the page, or check previous months.
My telescope
Comet McNaught
The Moon
M42 (Orion Nebula)


BOOK SALE - I'm selling off some scholarly books via Amazon Marketplace. Have a look!



Unmodified DSLR triumph

One of the current myths of astrophotography is that a DSLR can't photograph emission nebulae unless its IR filter has been replaced or removed.

For proof that that is not so, click here. (Or here.) Malcolm Park took a remarkable picture of the Horsehead with an unmodified Canon 30D. And he did it from just 10 miles from central London, using a broadband nebula filter. This picture, with technical details, is going to appear in my new book, DSLR Astrophotography.

--- ---

Short notes

Vista, yes, but not yet: Reportedly, Microsoft's leading programming tool. Visual Studio 2005, does not run under Windows Vista without an update. It's a security matter; Vista is much more careful about programs that try to disrupt the operating system. A patch has been issued, but the lesson learned is that Windows XP software does not necessarily run under Vista unless it's rather fastidiously designed.

I am happy to report that my own programs, so far, have run under Vista with no problems. But I'll let somebody else have all the initial problems with other software before I adopt Vista myself.

What a way to sell software: Econo-Soft.com offers Photoshop CS2 and other major packages for pennies on the dollar. But read the fine print. They're selling you a "backup copy" with no license and no activation code. That is, when you get it, you probably can't install it or run it.

They think what they're doing is legal. I don't.

More on the Astronomy Classified Ad Wars: The new free ad service, AstroClassifieds, has grown to nearly 10% the size Astromart (measuring by the number of ads) after just a few days of operation. Meanwhile, Astromart has started charging new users $12 for accounts, even though still describing the service as "free" on the opening page.

If you like News of the Weird, you'll also like Daily Snopes. The main job of Snopes is to investigate rumors and urban legends, but their daily news page is a funny collection of strange things from the world's newspapers.

--- ---

Credit card industry gets well-deserved flak

As a harsh critic of the credit card industry, I was amused and delighted by the following pair of public statements.

Jan. 25: Senator Chris Dodd, chairman, opening Senate Banking Committee hearings:

I would like to put the credit card industry ... on notice. If you currently engage in any business practice that you would be ashamed to discuss before this Committee, I would strongly encourage you to cease and desist that practice.

Jan. 18 (knowing the hearings were coming), a press release from Chase Credit Cards:

Chase also will eliminate the two-cycle method of calculating finance charges. "In our continuing review of customer feedback, we found that this practice was difficult to understand," added (Carter) Franke (chief marketing officer for Chase Card Services). "So, in the interest of clarity and transparency, we will utilize a simpler, one-cycle method of calculating finance charges." This change will reduce the finance charges for consumers when they choose not to pay for a new purchase in full. The elimination of two-cycle billing will be completed over the next few weeks.

That is, they're no longer going to charge you interest retroactively on debts you've already paid if you don't pay in full the next month.

The Government Accountability Office has issued a report highly critical of some present-day credit-industry practices. Text of full report here. That's the basis of the Senate hearings.

Some juicy testimony from Thursday's hearing is reported here. Choice quotes:

  • "No one has to be a crash test expert to buy a car. It's time for safety regulations in credit card products as well." — Elizabeth Warren (Harvard U.)
  • "The credit card is one of the only contracts in common law anywhere in which the superior bargaining entity can change its terms at anytime for any reason. If they can change the contract on an existing balance, then they will always have an unfair advantage." — Michael Donovan (Philadelphia)

[Note added 2011:] Let us not forget that the reason for all this nonsense is that credit card companies are not measuring the borrower's ability to pay. Instead, they are guessing, and then doing wild things when they guess wrong.

Along the way I learned that Springfield, Massachusetts, has a newspaper called The Republican which supports the Democrats.

Why that picture looks so good

The picture of my telescope that I posted 2 days ago has attracted some comment. How did I get the telescope to stand out so clearly from the background? Special lighting?

No; unsharp masking in Photoshop, more than once, with two or three different radii.

Unsharp masking is a technique that goes back to the Good Old Days of film. An "unsharp mask" is a blurry low-contrast negative image, made by contact-printing the original through a thin piece of ground glass or the like. Sandwich the unsharp mask with the original, and you get a picture with high contrast in the fine detail and lower contrast in the larger features. Make a copy, pumping up the contrast, and the fine details really jump out at you. This was standard practice with technical illustrations in the 1950s.

J# dying; will Java be next?

Java is a fine programming language, but I'm beginning to think it's a tiny bit past its heyday. Its great appeal is portability — write a program and run it on any computer, UNIX, Windows, or Macintosh. It is much better and safer than ActiveX for web client applications.

Most of the good ideas in Java were taken up in C#, which, of course, is not portable at all (until the Mono Project gets a bit further along).

And now Microsoft is discontinuing J#, which is its Java-like compiler for .NET with complete C# interoperability. Apparently the demand just isn't there.

Part of the problem with Java — which will hit C# too if it's not careful — has to do with versions. Many of the most useful tools still only support Java 1.4. But Sun is up to Java 1.6, and incompatibilities often arise.

--- ---

Telegrams are not dead yet

Of course, whenever we say something is not dead yet, we mean that it nearly is.

One year ago today, Western Union delivered its last telegram. Now they only do money transfers. They claim that money transfers have actually been their primary business since 1871.

But Americans can still and send telegrams through International Telegram, a Canadian company.

Of course, e-mail has completely displaced telegrams if you just want to get a message to someone. But there are still some laws that give special privileges to telegrams. For instance, if there is a deadline to cancel a contract, you can make the deadline by sending a telegram before the appointed time, regardless of when it is delivered. The telegram company's timestamp is legally valid.

There are also people who send telegrams to keep up traditions. It's a way to express condolences or congratulations without disturbing the people on the receiving end, who may not want phone calls and may not be checking e-mail.

I got a Western Union telegram on my first birthday in 1958; it's still packed away somewhere in this house. When my father died in 1966, we must have gotten a dozen telegrams of condolence. I got a Western Union Mailgram, not a telegram, to notify me in 1973 that I had been selected for a government-sponsored trip to Australia. I sent a telegram from England in 1976 to contact Yale Graduate School; in those days we didn't have fax, and I didn't want to spend $50 trying to reach someone by telephone.

And that was it. FedEx and Express Mail made it easy to send letters overnight. Telephony became cheaper, fax came in, and then e-mail. No more telegrams. Or, alternatively, you could say I send and receive dozens of them every day, through the Internet.

--- ---

What my telescope looks like

Most of the astronomical pictures that you see in this Notebook are taken with this equipment. The picture is in black-and-white because it's going to be a book illustration.

Piggyback astrophotography setup

The telescope is a Meade LX200, 8 inches (20 cm) in diameter, with an 8x50 finderscope. At the upper left you see my autoguider (for particulars click here and scroll down a bit); it watches the stars and corrects the tracking of the telescope to keep it perfect all the time.

Around the front of the telescope is a Kendrick dew heater (to prevent dew). Mounted "piggyback" on the telescope is a Canon 300-mm f/4 telephoto lens and a Canon XTi camera with a Canon Angle Finder C.

It takes me only 5 minutes to set up all of this on top of a permanently installed pier in my backyard that has an axis parallel to that of the earth. At dark-sky sites I have a portable tripod that takes its place, which I align by sighting stars.

Note that the pictures are taken through the telephoto lens, and the autoguider has its own lens. Often the telescope is not used at all, except to carry the rest of the equipment! Sometimes I photograph through the telescope.

Telephoto Moon

While setting up to take the picture above, I noticed the Moon high in the sky and decided to take a shot at it. This is an auto-exposed but manually focused picture of the Moon through the 300-mm telephoto lens (not the telescope). Is that a sharp lens, or what?


This is the central part of the picture, of course, and it was sharpened in Photoshop.

First light with the EOS 20Da


I finally got to try out the Canon EOS 20Da that Canon USA is lending me. Above is a stack of five 30-second exposures of the Orion Nebula in moonlight, minus two dark frames. (I know you think the Orion Nebula is the only thing I ever photograph, but it's good for testing equipment.)

The 20Da, now unfortunately discontinued, is the astronomy version of a Canon DSLR, and it has two superpowers:

  • Extended red sensitivity. As you can see in the picture, it picks up emission nebulae better than an ordinary DSLR because of better response to the wavelength of hydrogen-alpha.
  • Live focusing. This is handy! On the LCD screen, you can see a continuous, greatly magnified view of what the camera is aimed at. This enables me to focus perfectly on the stars in five or ten seconds — a process that could otherwise take many minutes.

It also looks as if this 20Da has a very high-grade sensor with few hot pixels.

I found out a bit too late that I don't have a cable release that fits this camera, so I couldn't take exposures longer than 30 seconds.

--- ---

Short astronomy notes

I'm too busy to write notebook entries, but here are a few things to look at:

Spectacular picture of Comet McNaught taken by its discoverer. This of course is not the discovery picture. When Robert McNaught discovered it, it was a 17th-magnitude speck.

Picture of 20 successive Full Moons. They are not the same; look at the subtle variations caused by the fact that the moon's orbit is not a perfect circle.

It looks like Astromart (which I no longer use) has a viable competitor. AstroClassifieds is a new, free classified-ad service for amateur astronomers, free even for book authors like me.

--- ---

My father's watch

Bulova self-winding watch I was surprised the other day to see a familiar face on the web site of WatchConcepts.com.

This Bulova self-winding watch is just like my father's. My mother gave it to him in the mid-1950s. I inherited it from him when he died in 1966, started wearing it regularly around 1968, and continue to wear it until Melody gave me a digital watch in 1980. So this is the watch I wore around the world in 1973, and to Cambridge in 1977-78. I continued wearing it for special occasions such as our wedding. It disappeared in 1985 when we moved into a new house; I don't know if it was stolen or merely packed up in my mother's or sister's personal effects.

It's a fully mechanical watch, of course. "Self-winding" means that as you move your arm around, you hear a slight whirring sound and a moving weight tightens up the mainspring.

Like all mechanical watches, this one had to be set every day if you wanted it to be accurate to the minute. Something we've forgotten about life in the 1960s is that it was common for different people's clocks to disagree by several minutes; if you had an appointment somewhere, you always arrived early and looked at the clock at your destination.

I found it useful to synchronize my watch with the school's bells (for class changes); this could be done with 1- or 2-second accuracy, first thing every morning, and I used to surprise people by closing my notebook and standing up about one second before the bell sounded, as if I had heard it coming.

Apparently, this was a popular model of watch, and in poor condition, it can be obtained on eBay for about $40. But WatchConcepts is correct to ask ten times that for one that is in perfect condition and has been checked out by a watchmaker. [No, I didn't buy it. In fact, they had sold it before I saw their ad.]

My father's airplane

B-17G 'Scorchy II' If you go to a hobby shop in America and buy the Revell model of a B-17 bomber, you get a model of a relatively generic-looking plane whose history I don't know.

But the German branch of Revell (actually a different company with rights to the same trademark) makes a model of the very plane that my father flew in (one of them, anyhow). Here's the box art, courtesy of Pete Albrecht.

I find it slightly amusing that the Germans prefer to build models of a plane that bombed Germany.

A discovery about named anchors in HTML

As everybody knows, web pages can contain named anchors, places marked within them, generally created with <a name="xyz">...</a> or with <p id="xyz">.

Then, to link to that particular position, you can do <a href="#xyz"> within the document or <a href="http://somewhere.com/filename.html#xyz"> from outside it.

That's exactly how the permalinks on these Notebook entries work.

Well... Whenever I opened one of my own files with Internet Explorer — that is, a web page existing as a file on my own computer rather than on a web server — the named anchors didn't work.

Strange bug in Internet Explorer? That's what I thought, because Opera Browser had no problems.

Then I found the answer. It has to do with how you open the file. You can open the same file either as:


or as:


And only the second of these is a URL and can take a named link (such as #xyz) after it. The first format, which Internet Explorer uses by default, is not a URL, and named links are not accepted.


Built-in Wi-Fi is better: It recently came to our attention that Melody's Toshiba laptop had a slot for a built-in Wi-Fi adapter, including antenna leads. So we bought one and put it in, removing the PC Card wireless adapter that was formerly used. Presto... signal strength went from "low" to "excellent" and there is no longer anything sticking out the PC Card slot threatening to get bumped and cause damage.

Classic detective stories: Some of the works of R. Austin Freeman have become available on Project Gutenberg. Yesterday, to take a break in between big projects, I read this one.

Freeman, you'll recall, invented the scientific detective. (Sherlock Holmes did a bit of scientific detecting, but it wasn't his main focus.) One of Freeman's best novels, The Red Thumb Mark, dealt with how to forge a fingerprint, back when fingerprint identification was a brand-new idea.

The EWD Papers are on the Web. These are E. W. Dijkstra's notes, mostly about the logic of computer programming, and they spearheaded a historic change in the nature of computer programming in the 1980s. Before Dijkstra, a computer program was a list of things to do; now it's a definition of how to do something. This change of perspective is absolutely vital if you want to write large programs without getting lost in your own mistakes.

--- ---

Buchwald's farewell

Two days ago (Jan. 17) the mail brought an issue of Reader's Digest containing a surprisingly cheerful, entertaining farewell essay written by humorist Art Buchwald, who was terminally ill with kidney failure.

The same day that the magazine was delivered nationwide, Mr. Buchwald died. He will be missed. I enjoyed reading his columns whenever I saw them. He often claimed to be writing about Washington from an office high in the top of the Washington Monument, commanding a good view of the city. Think of him as a more subtle, more dignified precursor of Dave Barry.

A cohort of women born around 1952

Somewhat in Buchwald's spirit, I want to wish a happy 55th birthday, this year, to the flight attendants of America.

The issue came up the other day when a Delta stewardess mistakenly said "Thank you for flying Western" (Western has been out of business for years) and then disclosed her age and that of her colleagues.

Back in my student days I remember noticing that most flight attendants were about 5 years older than me. They still are. A huge cohort of them started their careers around 1974, when the gasoline shortage got people started flying rather than driving. The domestic airlines haven't expanded much since then, so everybody's still on the job, with an average age that increases 1 year per year.

--- ---

Spammer convicted, headed for prison

From a Department of Justice news release:

Los Angeles, CA - In the nation's first jury conviction under the CAN-SPAM Act of 2003, an Azusa [California] man has been convicted of sending thousands of e-mails to America Online users that appeared to be from AOL's billing department and prompted the customers to send personal and credit card information, which he used to make unauthorized purchases.

In verdicts reached late Friday, Jeffrey Brett Goodin, 45, was found guilty of operating a sophisticated "phishing" scheme. The jury found that Goodin operated an Internet-based scheme designed to obtain personal and credit card information by tricking people into believing that they were providing information to a legitimate business.

The evidence presented during a week-long trial showed that Goodin used several compromised Earthlink accounts to send e-mails to AOL users. Those e-mails appeared to be from AOL's billing department and urged the users to "update" their AOL billing information or lose service. The e-mails referred the AOL customers to one of several webpages where the victims could input their personal and credit information. Goodin controlled those webpages, where he collected the information that allowed him and others to make unauthorized charges on the AOL users' credit or debit cards.

In addition to the CAN-SPAM Act conviction, Goodin was convicted of 10 other counts, including wire fraud, aiding and abetting the unauthorized use of an access device (credit card), possession of more than 15 unauthorized access devices, misuse of the AOL trademark, attempted witness harassment and failure to appear in court.

Goodin is scheduled to be sentenced by United States District Court Judge Christina A. Snyder on June 11. At sentencing, Goodin faces a statutory maximum sentence of 101 years in federal prison.

Definitely a good day for civilization. More detail here. Note that this was a conviction by a jury, not a no-show (default) or plea bargain.

He was arrested almost a year ago and presumably did his dastardly deeds in mid- or late 2005.

And that brings out one of the problems with computer crime: the time scale of justice. In my experience (and at one time, while handling security incidents for the University of Georgia, I got some experience), computer criminals think in terms of very short intervals. They are almost like shoplifters. They think that if they can get away with something for one week, it will never come back to haunt them.

About the only thing that can be done is put an appreciable number of these con artists in prison and keep reminding the public that they're there.

I would also resist any effort to legitimize or glamorize them as "hackers". Fraud, impersonation, and massive pestering are not feats of the intellect. I think the jury must have seen this very clearly. Spammers should always get jury trials.


It's just past midnight on the 18th and a minor ice storm is predicted — but I don't know whether to believe the predictions. Los Angeles and San Antonio have had disabling snow this week.

Here is a good collection of amusing and wise quotations about computer programming.

Need to disassemble a Toshiba laptop (as I did the other day)? Detailed instructions here.

--- ---

Why can't skeptics be skeptical?

It's Applied Epistemology Day in the Daily Notebook, so hang onto your hats.

Epistemology is the study of how we know things. And my concern today is how to tread the thin line between being gullible and being closed-minded.

Nobody wants to end up believing a lot of things that aren't true. So a healthy skepticism is justified when people report space aliens, ghosts, miracles, ESP, or whatnot.

But there is also an opposite error, which some people spend their time promoting. That is the error of assuming there can never be anything that isn't already in your belief-system. Obligatory "debunking," in other words.

The trouble with "debunking" is that it prevents discovery. As Cathy recently pointed out, when you set up debunkers as heroes, people believe anything debunkers say. The skeptics become blind believers. And then, if something strange does exist, they're never going to know.

For a more technical explanation of a logical error commonly made by debunkers, see Jeff Duntemann's recent article (to which I contributed).

--- ---

Stupid antivirus tricks

On Saturday evening, we wanted to back up some folders from Melody's laptop by simply copying them to a shared folder on another machine.

Simple, right? But slow. So we set the process going and ate dinner.

Unknown to us, the folders included some temporary files containing virus definitions downloaded by Norton Antivirus.

Well... When these copies of its own files arrived on the other machine, Norton (running there) reported "Intrusion detected!" and cut off communication with the laptop.

We can do without that. At least it wasn't HAL 9000 controlling my life support.

--- ---

Yes, I did see Comet McNaught

Uploaded early. Read tomorrow's news today!

Comet McNaught

Yes, I did see and photograph Comet McNaught back on January 9th, but the air in Georgia wasn't very clear. After work, right after sunset, I went to the roof of one of the University of Georgia's parking decks and set up my tripod and camera (Canon XTi, Sigma 105/2.8 lens). Icy air was blowing in my face and small clouds were zipping around the western sky.

The comet was a plain sight in 8x40 binoculars but was not visible to the unaided eye. On my pictures, it wasn't visible until I used Photoshop to increase the contrast and Neat Image to remove the grain. But it's bright; the picture you're looking at is a 1/125-second auto exposure.

--- ---


Two birthdays: Yesterday I should have wished happy birthday not only to HAL 9000, but also to my father-in-law, Jim Mauldin. I hasten to add that they are not related.

Fame thrust upon me: I was interviewed about HAL 9000 and made the front page of the Athens newspaper. (To read the article, you have to register online, but there is no charge.)

Things I've caught myself saying recently: "If you think I'm disagreeing with you, you're wrong." That's a one-sentence logic puzzle almost on the level of Raymond Smullyan's "I could not fail to disagree with you less."

Daylight comet: Pete Albrecht has managed to photograph Comet McNaught in full daylight, before sunset. (See his January 11, 2007, entry.) I saw the comet a couple of evenings ago and got some poor-quality photos. It's very bright but also very close to the sun.

Textbook payola scandal

I didn't believe it when the media first alleged that college professors were being paid kickbacks to require their students to buy particular textbooks.

But it has really happened and the Chronicle of Higher Education even found people willing to admit they'd taken the payola.

Folks, this is not a difficult ethical question. When an employer is paying you to do a job, you don't take money from a third party, without the employer's permission, to alter the way you do it.

It's perfectly OK for a professor to be paid (by a publisher) for writing a book, or reviewing one. But not for altering the way he conducts his course, nor for requiring the students to buy a particular book.

Justing from the Chronicle's interview, some professors need a bit of training in business ethics. That won't make them honest, of course, but it should at least make them realize they're doing something that others might object to.

As a state employee, I could be prosecuted if I took a bribe to influence the way I do my state government work. As I understand it, people who offer me bribes could also be prosecuted.

--- ---


Academic success: My research group has had an article accepted by Journal of Psychopharmacology (dealing with an application of some of our computerized brain-and-language tests). More news soon. Today's challenge: rounding up all 7 co-authors (by phone, e-mail, and fax) and getting them to sign forms. I think I am the only one who hasn't changed employer or location since the research was actually done.

The flood is over. My ISP, Sectorlink, has reinstated spam filtering and I am no longer having to deal with 200 pieces of obnoxious drivel per day. During the California trip, I had to check e-mail twice a day to keep the pile from becoming unmanageably large.

There is a small risk that I am not receiving all my legitimate mail, but then, there was also a risk of overlooking legitimate mail during the flood.

Happy birthday, HAL 9000. Real computers are not like that, thank goodness. In retrospect, one of the design goals of HAL 9000 must have been to fool its users into thinking that it had human-like intelligence.

Wooden pens: Everybody have a look at Pens and More, Melody's father's woodworking enterprise. This is where several of my best fountain pens came from.

Astrocamera: A Canon EOS 20Da has arrived, on loan from Canon Professional Services, and naturally the sky has clouded up. Pictures in a week or two...

Book review: I've revised the scathing book review that I wrote for Amazon the other day. (I don't know whether the new version will be up by the time you read this.) Parts of the book weren't quite as bad as I thought. But I also found more places where the author had obviously swallowed other people's hoaxes.

Meanwhile I stay very busy!

--- ---

Can't set Outlook Express as default mail client

Outlook Express is not your default mail client

The latest Windows Update has brought back this problem. The solution given at that link still works.

--- ---

How not to write history of science

As reading matter for the flight home, I bought the book Discarded Science: Ideas that Seemed Good at the Time, by John Grant. It's about strange obsolete theories of the past, or so the blurb indicates.

I didn't start reading it until I got back home, and then I wrote the scathing review that you can read here.

There's been plenty of strange science and pseudoscience in the past. A good book about it would be worth having. This isn't it.

--- ---

Congratulations, Melody!

Speaking of professions, Melody is now the publication designer for Electrostatic Spraying Systems. Yesterday was her first day. That's one reason we're busier than we used to be.


Want to understand nanotechnology? Click here.

This watch will soon be the subject of a Daily Notebook entry. And I don't even own it. Wait and see...

Latest word is that Canon is going to lend me an EOS 20Da to try out during my copious spare time.

Book success: At last the book industry seems to be emerging from its downward spiral. The first few months' sales of the latest Dictionary of Computer and Internet Terms were relatively good. This book is in every Barnes & Noble and Borders in the country. You need one!

--- ---

We don't work for free

[Revised for conciseness January 9.]

A note to the eager public: Melody and I don't practice our professions free of charge. Especially when we're as busy as we are right now.

We're glad to reply to quick questions by e-mail, but we can't handle hour-long phone calls or lengthy e-mail exchanges from strangers who want us to design something, fix something, write a computer program, or troubleshoot their equipment remotely. Unless, of course, they want to engage us at our usual hourly consulting rates and wait in line behind projects that are already under way.

All our pictures are copyrighted. We can and do sell reproduction rights, but we can't let you reproduce a picture free of charge. The equipment to take the pictures is not cheap, nor is the time to deal with requests.

(A note to our friends and family: This isn't directed at you. Total strangers have been asking us to do substantial amounts of work, right out of the blue. We can't.)

Speaking of professionals, Camera Tech Anaheim looks very promising. I was in there the other day. The owner has a remarkable stock of both knowledge and parts. I bought from him a replacement switch knob for my Mamiya/Sekor 500 DTL. (Where else can you buy 35-year-old camera parts right off the shelf?) I'm probably going to have him refurbish my Petri 7 (the first camera I used seriously, from 1967 to 1970) and I'll report back on the results. He specializes in a lot of things, one of which is the Petri 7.

The alert reader will note that I'm back in Georgia, although the combination of modern security measures and bad weather made the journey somewhat challenging. Los Angeles to Athens, Georgia, is almost exactly 2000 miles. The trip took 11.5 hours door to door. That's an average speed of 173.5 mph, only three times the speed of a car.

--- ---

Southern California

Saddleback Mountain from Anaheim Hilton

Notebook entries will continue to be scarce for the next several days or more, but here's an update. I've just been to sunny Southern California for the meeting of the Linguistic Society of America, where Cati Brown and I presented this piece of research and ended up being in the day's Language Log (a newsletter for linguists).

Briefly, we have developed a computer program which, when given a sample of the English language, measures the idea density, the extent to which pieces of information are connected together. Psycholinguists have been measuring idea density for a long time, but not by computer. We're going to save people some work.

It has been a very eventful long weekend. Only part of it will be chronicled here. I had many good visits with old friends.

The old neighborhood

Kahlua Apartments

As newlyweds from August 1982 to May 1984, Melody and I lived in this improbable-looking apartment building, the Kahlua Apartments, complete with wooden tiki gods, at 5339 North Rosemead Boulevard, San Gabriel, California (the building on the left in the picture). Immediately to the north and east is Temple City.

So how has the old neighborhood changed in 23 years? I went up there to find out. Melody and I had already made an expedition in January 1999, so what I found was not a total surprise, and this may not interest anyone but me, but here goes...

Many vacant lots and small empty areas have been filled in. The shopping center directly north of the apartments is still there, of course, and not greatly changed. The K-Mart looks very prosperous and has been repainted with lots of deep orange. Clark Drugs is now a CVS; the decor is completely modern but the type of store hasn't changed. (Enormous pharmacies seem to be a California thing.) Albertson's, the grocery store, is quite unchanged, not even redecorated, and now has a campy 1970s look.

Of course, to us, the place where we did our first grocery shopping as newlyweds is hallowed ground.

Continuing into Temple City, Casa del Rey, the Mexican restaurant, is still there and as good as ever. The northeast corner of Rosemead at Las Tunas is being cleared; I can't remember what was there before. (Houses?) [Melody says it was a movie theatre. Now I remember...]

A few blocks north on Rosemead, the store where we bought our first computer (a TI-99/4A) is now a discount grocery of some type. Much farther north, at Rosemead and Del Mar, the Naugles where we ate Macho Nachos late in the night has become a Del Taco.

Farther north, and making this more into a computer geek's tour...

Pasadena's "Radio Row" has changed. Dow Radio, a ham radio landmark since the 1930s, is no longer there; what was left of it was moved to Marvac Electronics in Costa Mesa. Fortunately, a new company, Du Vac Electronics, has set up a clean new store in the same location — the main store only; the rest of Dow's "row" extending down the block no longer has anything to do with electronics.

And C&H Sales Company, the surplus dealer, is closing its store after 60 years. I don't know if it's going to continue in business selling on eBay.

C&H Sales Company

What really propelled me to learn about electronics in our California days was the ready availability of inexpensive parts. You could get exactly what you needed — not just the Radio Shack subset — locally, often very cheap, without incurring any minimum orders or shipping expenses. Atlanta has never been like that. Thanks to Marvac, Costa Mesa still is.

At the time I didn't know whether all this dabbling in electronics would get me anywhere. I was supposed to be a linguist. But it worked out well... I may be a jack of all trades, but I'm fairly good at some of them.

--- ---

Camera history trivia

I've been extremely busy with other things, so this will be short, and it will be several days before I post anything more. But here are a couple of bits of camera history trivia.

The Olympus OM-1 is a work of genius, but it is actually a spinoff from an even more ambitious unfinished project — Maitani's Hasselblad-like modular 35-mm SLR. Follow the link, scroll to the middle of the article, and be amazed.

Why did the solid Japanese Topcon SLR of the 1960s have the same strange lens mount as the clunky East German Exakta? When shopping for my first SLR as a high-school student, that's one of many things I wondered about. Recently I found the answer (now I can't remember where): Medical instruments. Topcon makes more medical imaging instruments than cameras, and they originally designed them to fit the Exakta, which was the first widely available SLR.

Here is an enthusiast who goes into poetic raptures about unusual Nikon lenses. Notice especially one of the lenses that brought you Pearl Harbor.

--- ---

The book that launched Ramanujan

Carr's Synopsis

Every mathematician knows the story of Srinivasa Ramanujan (1887-1920) — the Indian clerk who taught himself mathematics from a British review textbook, started proving new theorems, wrote to G. H. Hardy at Cambridge, became a Fellow of Trinity, was recognized as a great mathematician, and then suffered failing health and died young.

The book that launched Ramanujan was Carr's Synopsis of Elementary Results in Pure and Applied Mathematics (1880). Recently, I tracked it down and borrowed Georgia State University's copy of it. Click on the picture above to see a PDF file of some representative pages. It's like that all the way through — theorems presented with little or no proof, one by one, such that a well-trained student might be able to work some of them out. It was definitely good brain food for Ramanujan.

While browsing through the file, chew on theorem 362 a moment. Lulled by the style of the book, I first thought it was something deep, and then I realized it is one of the most obvious things in all of arithmetic.

OM-1 makeover

I spent Friday afternoon working on my trusty 1976 Olympus OM-1. First I added a 1N34 diode in series with the battery, and a plastic ring inside the battery holder, so that it can use currently available MS76 and comparable batteries:

OM-1 diode installation

This was a wise move because it turned out the wire soldered to the battery holder was about to break (a common OM-1 affliction). After a few more rolls of film I would have had a camera with a dead meter.

Selecting the 1N34 diode: I was very successful with meter accuracy because I did two things. First, I ascertained that the typical current drain from the OM-1 meter circuit was about 0.5 mA under relatively bright-light conditions. Second, I selected a 1N34 whose voltage drop was as close as possible to 0.2 V at that current. One way to do the selection is to connect, in series, a 9-volt battery, a 22k resistor, and a diode, then measure the voltage across the diode.

Result (in my case): After modification, the meter agrees to within 1/2 stop with my vintage Olympus OM-2S. It's important to compare cameras from the same manufacturer because different companies use different calibration standards.

After completing the battery modification, I replaced the top cover because the poor camera had been hit on the head a few years ago. The bump did it no harm at all; there's about an eighth of an inch of empty space under the cover at that point. How many cameras with "impact damage" are actually perfectly OK, I wonder?

Here are "before" and "after" pictures. The new cover was supplied, with helpful advice, by John Hermanson.

OM-1 before OM-1 after

And here's how I'm protecting my fabulous-1970s camera from its fabulous-1970s clip-on strap:

OM-1 strap

A couple of pieces of vinyl tubing do the trick.

But while replacing the top cover I made a sad discovery. When I had an Atlanta camera shop replace all the foam in the camera about 10 years ago, they didn't get the foam around the prism. And Olympus unfortunately used a kind of foam that, after a quarter century or so, turns to goo and attacks the surface of the prism. Accordingly, when you look at the viewfinder you see a bit of black fuzz near the bottom of the frame, and it's not removable.

The problem and preventive measures are documented here. I removed all of the old foam. Later I plan to replace the prism. On eBay I found a $21 jammed OM-G that can serve as an organ donor. That's going to be a project for mid-January.

Olympus OM-10 sticky frame counter

I also fixed Cathy's Olympus OM-10, a camera that was given to her several years ago, and which she never used much because it almost immediately developed a quirk — the frame counter (film counter) wouldn't reset when the camera was opened, and we were afraid to run it forward beyond its normal range.

OM-10 frame counter

Removing the top cover of an OM-10 is surprisingly easy, much simpler than the OM-F-based procedure I wrote up about a decade ago:

  • Open the back of the camera and use tape to keep it from latching shut again.
  • Film speed dial: Take note of its setting and leave it there.
  • Film advance lever: Unscrew the black knob. Either use a spanner in its two holes, or (preferably) just press down on it with a big rubber stopper and turn counterclockwise.
  • Rewind knob: Hold the shaft so it doesn't rotate, and unscrew the knob (counterlockwise). When holding the shaft, take care not to damage it; fitting a piece of wood between the prongs is one good way to hold it.
  • On-off switch: Remove the C-clip around the rewind shaft. The on/off switch lifts out.
  • Six screws around edges: Self-explanatory.
  • Spring: Like all Olympuses, the door spring is at the front left of the camera and will probably come loose; it may even jump across the room. Take note when lifting the cover.
  • Reassembly: Note orientation of shutter release button so that it will fit into the hole in the top cover. Insert the door spring carefully; if necessary, it can be adjusted, or even inserted, through the on/off switch opening after the top cover is on.

Now for the story of the frame counter. As you can see in the picture, it consists of a plastic gear wheel with a numbered ring glued to the top, held in place with a C-clip. Beneath it is a spring that rewinds it to "S" whenever it's released, and to the left is a ratchet mechanism that advances and holds it as you take pictures.

Thank goodness there wasn't anything wrong with the ratchet mechanism or anything farther down the shaft. The problem was simply that the plastic wheel didn't turn freely all the time. It was intermittent; I could loosen it up, and then it would stick again.

The problem was completely cured by removing the C-clip. That showed me the wheel wasn't sticking on the shaft or being gripped by anything I didn't know about. But the C-clip obviously has to go back on, so this mechanism doesn't take itself apart while Cathy is taking pictures.

I took a good look at the plastic gear and noticed that it seems to have come out of the mold with a bit more material than intended — it bulged. I measured its thickness at 2.7 mm.

Figuring the maker probably intended 2.5 mm, not 2.7, I sanded the wheel down to that thickness using fine and then superfine sandpaper. In the picture you see it after that treatment; the top and bottom surfaces were slightly bulgy and irregular beforehand. Now the C-clip goes on easily into its slot on the axle without binding the plastic wheel.


I should add that I am not a camera repair expert and have never attempted anything complicated, such as adjusting a shutter. But some repairs are simple, and the activity, though time-consuming, is better than watching ball games or building ships in bottles. I learned all I know from Tomosy's books.


First dubious achievement award of 2007: To the U.S. Postal Service for closing on January 2 in honor of Gerald Ford. Nothing against President Ford, but we weren't expecting 2 postal holidays on consecutive days, and we're awaiting some important mail.

End of the credit-card era? Bankrate reports that almost nobody wants new credit cards any more. The response rate to direct-mail offers of new credit cards is down to 0.03% (one in 3300), from 2.2% (one in 45) in 1993.

As for the worst deal in the business, it may be the "Imagine MasterCard" referenced in the link I just gave you. Read the terms here: $150 annual fee and $119.40 per year of maintenance fees, all charged directly to your checking account. All this for a $350 credit limit and a nearly 20% interest rate. Why would anyone want that credit card? I suppose it's for people who imagine they're creditworthy.

Want to calibrate your monitor for correct color? Click here.

Funniest machine of 1940: Possibly this gadget for playing music with a car's exhaust.

--- ---
--- --- --- --- ---
If what you are looking for is not here, please look at previous months.